ORIGINAL ARTICLE

Expression of matrix metalloproteinases in patients withbipolar disorderFabria Chiarani,1,2 Gabriel Rodrigo Fries,1,2 Laura Stertz,1,2 Keila Maria Cereser,1 Angela T. S. Wyse,2

Flavio Pereira Kapczinski,1,2 Maurıcio Kunz1

1Molecular Psychiatry Unit and National Science and Technology Institute for Translational Medicine (INCT-TM), Hospital de Clınicas de Porto

Alegre (HCPA), Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, RS, Brazil. 2Graduate Program in Biological Sciences,

Biochemistry, UFRGS, Porto Alegre, RS, Brazil.

Objective: High cardiovascular mortality rates have been reported in patients with bipolar disorder(BD). Studies indicate that matrix metalloproteinases (MMPs) are implicated in cardiovasculardiseases. We evaluated the expression pattern of MMP-2 and MMP-9 in blood from patients with BDduring acute mania and after euthymia, in comparison with healthy controls.Methods: Twenty patients and 20 controls were recruited and matched for sex and age. MMPmessenger RNA (mRNA) levels were measured using real-time quantitative polymerase chainreaction (PCR). Body mass index (BMI) was calculated for all subjects.Results: There were no significant differences in MMP-2 and MMP-9 mRNA expression betweenpatients and controls. mRNA levels were not significantly different during mania and euthymia.However, MMP-2 mRNA levels were negatively associated with BMI in BD patients and positivelyassociated with BMI in controls. There was no difference in the pattern of MMP-9 expression betweenpatients and controls.Conclusions: Our results suggest a different pattern of association between MMP-2 and BMI in BDpatients as compared with controls. Despite some study limitations, we believe that the role of MMPsin BD should be further investigated to elucidate its relationship with cardiovascular risk.

Keywords: Bipolar disorder; matrix metalloproteinases; cardiovascular disease; mania;euthymia

Introduction

Bipolar disorder (BD) is a mood disorder associated withcyclic episodes of mania and depression. BD is estimatedto affect 4% of the population,1 at a significant health carecost to society. It is also associated with high comorbidityrates, which, in turn, are associated with other generalmedical conditions.2

Adults with BD are at increased risk of prematurecardiovascular disease, which is prevalent over a decadeearlier than in the average population.3 BD has beenassociated with high prevalence of cardiovascular dis-ease (about 49%),4 and patients with BD experiencetwice the cardiovascular mortality expected from generalpopulation estimates.5 After suicide and accidents,cardiovascular and vascular diseases are the leadingcauses of death in this patient population.5

A prospective cohort study of participants with BDshowed that (hypo)manic symptoms independently pre-dicted cardiovascular mortality; the results also showedthat mood disorders may mediate vascular disease, and

that the mechanisms associated with this mediation ofvascular disease are independent of and distinct fromtraditional risk factors.6 Moreover, alterations in cardiacvariability and predictability in BD may be state-depen-dent.7 Although several studies have investigated themechanisms that underlie high cardiovascular risk in BDpatients, these mechanisms are not fully understood.

Matrix metalloproteinases (MMPs) are a group ofendopeptidases produced by the structural componentsof the vascular wall and inflammatory cells, which havethe capacity to cleave several components of theextracellular matrix (ECM). MMP-2 and MMP-9 cleavedifferent bioactive molecules, including cytokines, che-mokines, cell surface receptors, and growth factors,8

contributing significantly to the pathogenesis of severalcardiovascular diseases, such as atherosclerosis, hyper-tension, heart failure, and ischemic heart diseases.9,10

Pathological processes such as inflammation andoxidative or nitrosative stress play an important role inthe expression, secretion, and activation of MMPs.11

Thus, oxidative and inflammatory pathways may explainsome of the clinical features of BD.12

In view of the foregoing, this study aimed to evaluatethe expression of MMP-2 and MMP-9 messenger RNA(mRNA) in the blood of BD patients during manicepisodes and after achievement of euthymia in compar-ison with healthy subjects.

Correspondence: Fabria Chiarani, Hospital de Clınicas de PortoAlegre, Laboratorio de Psiquiatria Molecular, Rua Ramiro Barcelos,2350, CEP 90035-903, Porto Alegre, RS, Brazil.E-mail: fabriachiarani@yahoo.com.brSubmitted Sep 21 2012, accepted Jan 16 2013.

Revista Brasileira de Psiquiatria. 2013;35:375–379� 2013 Associacao Brasileira de Psiquiatria

doi:10.1590/1516-4446-2012-1004

Methods

Patients with BD were recruited from Hospital de Clınicasde Porto Alegre (HCPA), Porto Alegre, Brazil. Allparticipants underwent a comprehensive clinical interviewby a psychiatrist. The diagnosis of BD was established onthe basis of the available clinical information andconfirmed with a Structured Clinical Interview for DSM-IV –– Axis I (SCID-I) and a standard protocol to assesspatient psychopathology and clinical features. Baselineassessment, including blood sampling, was performed atadmission to the psychiatric unit. Patients had to fulfillDSM-IV criteria for a manic episode for inclusion in thestudy. All patients received naturalistic treatment at thediscretion of the attending psychiatrist and had nosignificant comorbid medical conditions. Mood symptomswere assessed using the Young Mania Rating Scale(YMRS)13 and the Hamilton Depression Rating Scale(HAM-D), 21-item version.14 A second assessment wascarried out after remission of manic symptoms. Patientswere considered euthymic if they presented scores , 7on both the YMRS and HAM-D scales. BMI was definedas weight divided by height squared (kg/m2).

Healthy controls had no personal history of neurologi-cal or psychiatric illness and no first-degree relatives withpsychiatric disorders. Patients and controls werematched for age and sex and were free of clinicallyapparent cardiovascular diseases. The assessment ofcardiovascular status was based on the patient’s clinicalhistory (self-reported data and data obtained frommedical records) and no intake of medications used forthe treatment of cardiovascular diseases.

Twenty patients and 20 healthy controls (10 males and10 females in each group) participated in the study. Meanage was 38.3563.33 years in the BD group and37.0763.64 years in the control group. All patients weretaking at least one mood stabilizer (lithium, valproic acid)and one antipsychotic. Fifty percent of the patients werebeing treated with first-generation antipsychotics (halo-peridol and chlorpromazine), and the others with atypicalantipsychotics (risperidone, olanzapine, and clozapine).

The study was approved by the local EthicsCommittee, and informed consent was obtained from allparticipants prior to their inclusion in the study.

RNA extraction and cDNA synthesis

Peripheral blood was collected from patients and healthyvolunteers in PAXGene Blood RNA Tubes (PreAnalytiX).Total RNA was isolated from whole blood using thePAXGene Blood RNA Kit (Qiagen) according to manu-facturer instructions. RNA concentrations were assessedusing NanoDrop (Uniscience) equipment. RNA wasconverted to cDNA using the High-Capacity cDNAReverse Transcription Kit (Applied Biosystems) in a totalreaction volume of 10 mL containing 2 mL of 10x RTbuffer, 0.8 mL of 25x dNMT Mix (100 mM), 2 mL of 10x RTRandom Primers, and 1 mL of MultiScribe ReverseTranscriptase (50 units/mL) (all according to manufac-turer instructions). Reactions were carried out for 10 minat 256C, 2 h at 376C, and 5 s at 856C. Subsequently,

cDNA was kept at -206C until it was used for polymerasechain reaction (PCR) amplification. MMP-2 and MMP-9gene expression were measured by real-time quantitativereverse transcription-PCR (RT-PCR) using specificTaqMan FAM/MGB assays (Applied Biosystems, IDassay Hs01548727_m1 for MMP-2 andHs00234579_m1 for MMP-9).

Expression values were normalized by b2M endogen-ous control expression using a TaqMan VIC/MGBendogenous control inventoried assay (AppliedBiosystems, 4326319E). Reactions were performed inan Applied Biosystems 7500 Real-Time PCR System,which detects the PCR product directly without down-stream processing. Reactions were carried out in a totalvolume of 12 mL with 6 mL of 2x TaqMan GeneExpression Master Mix (containing ROX, Amplitaq GoldDNA polimerase, AmpErase UNG, dATP, dCTP, dGTP,dUTP, and MgCl2), 0.6 mL of 20x TaqMan GeneExpression Assay, 0.6 mL of 20x TaqMan EndogenousControl, 3.8 mL of water, and 1 mL of cDNA solution. Thecycling program consisted of 2 min at 506C and 10 min at956C followed by 40 cycles of 15 s at 956C and 1 min at606C. All reactions were performed in triplicate. Relativeexpression levels were determined by the ddCt methodas described by Livak & Schmittgen.15

Statistical analysis

Statistical analyses were performed using SPSS version18.0. Results were considered significant when p , 0.05.Normality tests were performed using the Shapiro-Wilkstest. All comparisons between and within BD patientswere performed using generalized estimating equations(GEE). Bonferroni’s test was used for post-hoc analysisof the data. The gamma distribution was used due to theasymmetric profile of variables. BMI was included as acovariate in a regression model, and results arepresented with and without the BMI factor. Data areexpressed as mean and standard deviation (SD) or asmedian and interquartile range (IQR).

Results

The BMI of BD patients and controls was not statisticallydifferent, with a median BMI of 27.2 (19.5-58.9) forpatients and 24.2 (17.1-31.6) for controls.

Median time between acute episodes and remission ofsymptoms in BD patients was 29.50 (20.50-58.00) days.The median HAM-D and YMRS scores in manic statewere 3.50 (1.75-5.00) and 27.50 (19.00-30.70), respec-tively, vs. 1.00 (0.00-2.50) and 1.00 (0.00-4.00) ineuthymia.

Table 1 shows the mRNA levels from BD patients andcontrols. MMP-2 and MMP-9 mRNA levels were notsignificantly different between BD patients and controls (p= 0.933 and p = 0.268, respectively). However, whenadjusted for BMI, the expression pattern of MMP-2 wasdifferent between BD patients and controls (p = 0.004).Our results showed a negative association between BMIand MMP-2 mRNA levels (Figure 1). There was no

F Chiarani et al.376

Rev Bras Psiquiatr. 2013;35(4)

difference in the expression pattern of MMP-9 betweenpatients and controls when adjusted for BMI (p = 0.483)(Figure 2). MMP mRNA levels did not differ between themanic and euthymic states in BD patients.

Discussion

In this study, the expression pattern of MMP-2 wasdifferent between BD patients and controls when adjustedfor BMI. The different MMP-2 mRNA levels found heremay explain the connection between BD and somaticillnesses such as cardiovascular disease.

Tissue remodeling and inflammation are central to theprocess of cardiovascular diseases both in acute events,such as myocardial infarction and sudden death, and inchronic conditions, such as vascular atherosclerosis andheart failure.16 Several MMPs have been implicated intissue remodeling, and the equilibrium between MMPsand their endogenous inhibitors is critical for the main-tenance of the integrity of the cardiovascular system.17

Inflammatory abnormalities have been associated withBD, and increased levels of inflammatory cytokines,particularly tumor necrosis factor-alpha (TNF-a) may playa critical role in BD.18-20 Shifts in mood states are a keyfeature of BD. It has been suggested that immune systemactivation may vary across affective states.21

Furthermore, oxidative stress plays a critical role in thepathophysiology of BD.22 The formation of reactiveoxygen species as well as the induction of signalingmolecules, such as cytokines, cause the activation ofMMPs,23 which may represent a possible mechanismlinking BD to the biological activation of MMP.

Recent studies have shown that patients with BD havea significant preponderance of the T allele vs. C allele ofthe 1562C/T polymorphism of the MMP-9 gene, suggest-ing that this polymorphism may mediate the associationbetween BD and cardiovascular disease.24,25 However,to our knowledge, this is the first study to report theexpression of the MMP2 and MMP9 genes in whole bloodof patients with BD, which is a group at increased risk ofcardiovascular disease. Although mean BMI in oursample was not different between BD patients andcontrols, when we consider MMP-2 mRNA levelsindividually in relation to BMI, the expression patternappears to be influenced differently by BMI. The resultsshow that MMP-2 gene transcription changes as afunction of BMI in BD patients, indicating that the profileof expression in patients is the opposite of that in controls.The influence of BMI suggests that MMP-2 expression intotal blood is dependent on obesity status, corroboratingthe hypothesis that adipose tissue, in addition to storingenergy, is an important endocrine organ that producesand secretes a variety of inflammatory cytokines,hormones, and other metabolic factors involved in thepathogenesis of atherosclerosis.26

Downregulation of MMP-2 activity has been demon-strated in premenopausal obese women, and is asso-ciated with attenuated ECM degradation andabnormalities of left ventricular function.27 AlthoughMMP-2 and MMP-9 are similar enzymes, our resultscorroborate a study that showed distinct promoter

Table 1 Matrix metalloproteinase messenger RNA levels for patients with bipolar disorder and controls

Bipolar disorder

Mania Euthymia Controls p-value Adjusted p-value*

MMP-2 1.18360.410 1.00060.367 1.02660.218 0.933 0.004{

MMP-9 0.87960.232 0.81760.267 1.39960.302 0.268 0.526

MMP = matrix metalloproteinase.Results expressed as mean 6 standard deviation.* Adjusted by body mass index.{ Difference significant for p , 0.05.

Figure 1 Effect of body mass index (BMI) on matrixmetalloproteinase 2 expression in patients with bipolardisorder (during mania and euthymia) and controls

Figure 2 Effect of body mass index (BMI) on matrixmetalloproteinase 9 expression in patients with bipolardisorder (during mania and euthymia) and controls

Matrix metalloproteinases in bipolar disorder 377

Rev Bras Psiquiatr. 2013;35(4)

elements and different spatial and temporal expressionpatterns of these proteases during ischemia/reperfusioninjury.11 It is possible that the type of upstream stimulioperative in BD will ultimately determine the transcrip-tional activity of selected MMP types.

Nonpharmacologic and pharmacologic factors alikecontribute to higher BMI in mental illnesses. In BDpatients, atypical depression, eating habits, behavior,physical inactivity, and lower basal metabolic rate duringdepression can influence the BMI.28

Some potential limitations of the present study includethe small number of subjects, the lack of information aboutlifestyle factors (e.g., eating habits and physical activity),the heterogeneous of time to remission of symptoms, andthe interference of drugs used by BD patients in theexpression pattern of MMPs. Studies have shown thatsome antidepressants and mood stabilizers may regulatethe expression and activity of MMP-2 and MMP-9 and oftissue inhibitors of the metalloproteinases (TIMPs) 1-4.29,30

An interesting future research direction would be studyof the gene expression of major MMP-9 and MMP-2inhibitors (TIMPs) and evaluation of MMP-9/TIMP-1 andMMP-2/TIMP-2 ratios. In a study by Rybakowski et al.,31

increased levels of serum MMP-9 in bipolar illness wereobserved only in depressive episodes, and not in manicones. Further studies should also be carried out to assessthe circulating levels of MMP-2 in BD patients.

Belo et al.32 found lower pro-MMP-2/TIMP-2 ratios inobese children and adolescents when compared to non-obese controls; conversely, the pro-MMP-9/TIMP-1 ratioswere comparable in these groups. Patients with essentialhypertension exhibit decreased MMP-2 and MMP-9plasma levels and activity; this may reflect abnormalECM metabolism. The increase in ECM depositioncomponents on the arterial wall facilitates vascularfibrosis.33 Derosa et al.34 demonstrated that both MMP-2 and MMP-9 plasma levels are significantly higher inobese patients than in controls. The decrease in MMP-2mRNA in response to a BMI increase may reflectfeedback inhibition of MMP-2 transcription by theincrease in circulating active MMP-2.35

Our study suggests a different pattern of expression ofMMP-2 associated with BMI in patients with BD incomparison to healthy controls. The study contributes tounderstanding the association between inflammation andclinical comorbidity in BD. Nevertheless, further trials arenecessary to clarify the relationship between MMPs andcardiovascular risk in BD. In addition, improved thera-peutic methods are needed to address the effects ofobesity on individuals with this mental illness. Theinvolvement of adipose tissue and immune systemabnormalities appears to be a potential biological targetthat will likely contribute to a better understanding of themechanisms involved in BD pathophysiology.

Acknowledgements

This research was supported by Fundo de Incentivo aPesquisa –– Hospital de Clınicas de Porto Alegre (FIPE-HCPA).

Disclosure

The authors report no conflicts of interest.

References

1 Merikangas KR, Jin R, He JP, Kessler RC, Lee S, Sampson NA, et al.Prevalence and correlates of bipolar spectrum disorder in the worldmental health survey initiative. Arch Gen Psychiatry. 2011;68:241-51.

2 Guo JJ, Keck PE Jr, Li H, Jang R, Kelton CM. Treatment costs andhealth care utilization for patients with bipolar disorder in a largemanaged care population. Value Health. 2008;11:416-23.

3 Goldstein BI, Fagiolini A, Houck P, Kupfer DJ. Cardiovasculardisease and hypertension among adults with bipolar I disorder in theUnited States. Bipolar Disord. 2009;11:657-62.

4 Fenn HH, Bauer MS, Altshuler L, Evans DR, Williford WO, KilbourneAM, et al. Medical comorbidity and health-related quality of life inbipolar disorder across the adult age span. J Affect Disord.2005;86:47-60.

5 Osby U, Brand L, Correia N, Ekbom A, Sparen P. Excess mortality inbipolar and unipolar disorder in Sweden. Arch Gen Psychiatry.2001;58:844-50

6 Fiedorowicz JG, Solomon DA, Endicott J, Leon AC, Li C, Rice JP,et al. Manic/hypomanic symptom burden and cardiovascularmortality in bipolar disorder. Psychosom Med. 2009;71:598-606.

7 Henry BL, Minassian A, Paulus MP, Geyer MA, Perry W. Heart ratevariability in bipolar mania and schizophrenia. J Psychiatr Res.2010;44:168-76.

8 Back M, Ketelhuth DF, Agewall S. Matrix Metalloproteinases inAtherothrombosis. Prog Cardiovasc Dis. 2010;52:410-28.

9 Castro MM, Kandasamy AD, Youssef N, Schulz R. Matrix metallo-proteinase inhibitor properties of tetracyclines: therapeutic potentialin cardiovascular diseases. Pharmacol Res. 2011;64:551-60.

10 Fontana V, Silva PS, Gerlach RF, Tanus-Santos JE. Circulatingmatrix metalloproteinases and their inhibitors in hypertension. ClinChim Acta. 2012;413:656-62.

11 Dejonckheere E, Vandenbroucke RE, Libert C. Matrix metallopro-teinases as drug targets in ischemia/reperfusion injury. Drug DiscovToday. 2011;16:762-78.

12 Berk M, Kapczinski F, Andreazza AC, Dean OM, Giorlando F, MaesM, et al. Pathways underlying neuroprogression in bipolar disorder:focus on inflammation, oxidative stress and neurotrophic factors.Neurosci Biobehav Rev. 2011;35:804-17.

13 Huang TL, Lee CT, Liu YL. Serum brain-derived neurotrophic factorlevels in patients with major depression: effects of antidepressants. JPsychiatr Res. 2007;42:521––5.

14 Lyoo IK, Sung YH, Dager SR, Friedman SD, Lee JY, Kim SJ, et al.Regional cerebral cortical thinning in bipolar disorder. Bipolar Disord.2006;8:65-74.

15 Livak KL, Schmittgen TD. Analysis of relative gene expression datausing real-time quantitative PCR and the 2(-Delta Delta C(T))method. Methods. 2001;25:402-8.

16 Liu P, Sun M, Sader S. Matrix metalloproteinases in cardiovasculardisease. Can J Cardiol. 2006;22:25B-30B.

17 Beaudeux JL, Giral P, Bruckert E, Foglietti MJ, Chapman MJ. Matrixmetalloproteinases, inflammation and atherosclerosis: therapeuticperspectives. Clin Chem Lab Med. 2004;42:121-31.

18 Brietzke E, Kapczinski F. TNF-a as a molecular target in bipolardisorder. Prog Neuropsychopharmacol Biol Psychiatry. 2008;32:1355-61.

19 Drexhage RC, Knijff EM, Padmos RC, Heul-Nieuwenhuijzen L,Beumer W, Versnel MA, et al. The mononuclear phagocyte systemand its cytokine inflammatory networks in schizophrenia and bipolardisorder. Expert Rev Neurother. 2010;10:59-76.

20 Kapczinski F, Dal-Pizzol F, Teixeira AL, Magalhaes PV, Kauer-Sant’Anna M, Klamt F, et al. Peripheral biomarkers and illnessactivity in bipolar disorder. J Psychiatr Res. 2011;45:156-61.

21 Brietzke E, Stertz L, Fernandes BS, Kauer-Sant’anna M,Mascarenhas M, Escosteguy Vargas A, et al. Comparison ofcytokine levels in depressed, manic and euthymic patients withbipolar disorder. J Affect Disord. 2009;116:214-7.

F Chiarani et al.378

Rev Bras Psiquiatr. 2013;35(4)

22 Steckert AV, Valvassori SS, Moretti M, Dal-Pizzol F, Quevedo J.Role of oxidative stress in the pathophysiology of bipolar disorder.Neurochem Res. 2010;35:1295-301.

23 Spinale FG. Myocardial matrix remodeling and the matrix metallo-proteinases: influence on cardiac form and function. Physiol Rev.2007;87:1285-342.

24 Rybakowski JK, Skibinska M, Leszczynska-Rodziewicz A,Kaczmarek L, Hauser J. Matrix metalloproteinase-9 gene and bipolarmood disorder. Neuromolecular Med. 2009;11:128-32.

25 Rybakowski JK. Matrix Metalloproteinase-9 (MMP9) - A MediatingEnzyme in Cardiovascular Disease, Cancer, and NeuropsychiatricDisorders. Cardiovasc Psychiatry Neurol. 2009;2009:904836.

26 Hajer GR, van Haeften TW, Visseren FL. Adipose tissue dysfunctionin obesity, diabetes, and vascular diseases. Eur Heart J. 2008;29:2959-71.

27 Kosmala W, Plaksej R, Przewlocka-Kosmala M, Kuliczkowska-Plaksej J, Bednarek-Tupikowska G, Mazurek W. Matrix metallopro-teinases 2 and 9 and their tissue inhibitors 1 and 2 in premenopausalobese women: relationship to cardiac function. Int J Obes (Lond).2008;32:763-71.

28 Megna JL, Schwartz TL, Siddiqui UA, Herrera Rojas M. Obesity inadults with serious and persistent mental illness: a review ofpostulated mechanisms and current interventions. Ann ClinPsychiatry. 2011;23:131-40.

29 Benekareddy M, Mehrotra P, Kulkarni VA, Ramakrishnan P,Dias BG, Vaidya VA. Antidepressant treatments regulate matrix

metalloproteinases-2 and -9 (MMP-2/MMP-9) and tissue inhibitors ofthe metalloproteinases (TIMPS 1-4) in the adult rat hippocampus.Synapse. 2008;62:590-600.

30 Tsai LK, Leng Y, Wang Z, Leeds P, Chuang DM. The moodstabilizers valproic acid and lithium enhance mesenchymal stem cellmigration via distinct mechanisms. Neuropsychopharmacology.2010;35:2225-37.

31 Rybakowski JK, Remlinger-Molenda A, Czech-Kucharska A,Wojcicka M, Michalak M, Losy J. Increased serum matrix metallo-proteinase-9 (MMP-9) levels in young patients during bipolardepression. J Affect Disord. 2013;146:286-9.

32 Belo VA, Souza-Costa DC, Lana CM, Caputo FL, Marcaccini AM,Gerlach RF, et al. Assessment of matrix metalloproteinase (MMP)-2,MMP-8, MMP-9, and their inhibitors, the tissue inhibitors ofmetalloproteinase (TIMP)-1 and TIMP-2 in obese children andadolescents. Clin Biochem. 2009;42:984-90.

33 Zervoudaki A, Economou E, Pitsavos C, Vasiliadou K, Aggeli C,Tsioufis K, et al. The effect of Ca2+ channel antagonists on plasmaconcentrations of matrix metalloproteinase-2 and -9 in essentialhypertension. Am J Hypertens. 2004;17:273-6.

34 Derosa G, Ferrari I, D’Angelo A, Tinelli C, Salvadeo SA, Ciccarelli L,et al. Matrix metalloproteinase-2 and -9 levels in obese patients.Endothelium. 2008;15:219-24.

35 Sternlicht MD, Werb Z. How matrix metalloproteinasesregulate cell behavior. Annu Rev Cell Dev Biol. 2001;17:463-516.

Matrix metalloproteinases in bipolar disorder 379

Rev Bras Psiquiatr. 2013;35(4)